Yes I have matched all the arguments. I should mention that the code
compiles and runs flawlessly using MPICH2-1.0.7 so it's got to be an
issue with my specific build of openMPI. I want to get openMPI up and
running for performance comparisons.
On Mon, Sep 22, 2008 at 6:43 PM, Jeff Squyres <jsquyres_at_[hidden]> wrote:
> What's the source code in question, then? Did you match all the arguments?
> On Sep 22, 2008, at 8:36 PM, Brian Harker wrote:
>> Nope, no user-defined types or arrays greater than 2 dimensions.
>> On Mon, Sep 22, 2008 at 6:24 PM, Jeff Squyres <jsquyres_at_[hidden]> wrote:
>>> On Sep 22, 2008, at 6:48 PM, Brian Harker wrote:
>>>> when I compile my production code, I get:
>>>> fortcom: Error: driver.f90: line 211: There is no matching specific
>>>> subroutine for this generic subroutine call. [MPI_SEND]
>>>> Seems odd that it would spit up on MPI_SEND, but has no problem with
>>>> MPI_RECV... What do you guys think? And thanks again for your help
>>>> and patience?
>>> The F90 MPI bindings have some well-known design flaws (i.e., problems
>>> the standard itself, not any particular implementation). Many of them
>>> center around the fact that F90 is a strongly-typed language. See this
>>> paper for some details:
>>> Here's the highlights, as they pertain to writing F90 MPI apps:
>>> - There is no equivalent to C's (void*). This means that the F90 MPI
>>> bindings cannot accept user-defined datatypes.
>>> - This also means that *every* pre-defined type must have a F90 MPI
>>> There are approximately 15 intrinsic size/type combinations. There are
>>> MPI functions that take one choice buffer (e.g., MPI_SEND, etc.), and 25
>>> functions that take two choice buffers (e.g., MPI_REDUCE). I'm copying
>>> math from the paper, and I think we got it slightly wrong (there was a
>>> discussion about it on this list a while ago), but it results in many
>>> *millions* of F90 MPI bindings functions. There's no compiler on the
>>> than can handle all of these in a single F90 module.
>>> Open MPI compensates for this with the following:
>>> - F90 bindings are not created for any of the 2-choice-buffer functions
>>> - F90 bindings are created for all the 1-choice-buffer functions, but
>>> for dimensions up to N dimensions (N defaults to 4, IIRC). You can
>>> the value of N with OMPI's configure script; use the
>>> --with-f90-max-array-dim. The maximum value of N is 7.
>>> So -- your app failed to compile because you either used a user-defined
>>> datatype or you used an array with a dimension greater than 4. If it was
>>> greater-dimension issue, you can reconfigure/recompile/reinstall OMPI
>>> (again, sorry) with a larger N value. If it was a user-defined datatype,
>>> you unfortunately have to "include mpif.h" in that
>>> subroutine/function/whatever, sorry (and you lose the type checking).
>>> Here's some info from OMPI's README:
>>> - The Fortran 90 MPI bindings can now be built in one of three sizes
>>> using --with-mpi-f90-size=SIZE (see description below). These sizes
>>> reflect the number of MPI functions included in the "mpi" Fortran 90
>>> module and therefore which functions will be subject to strict type
>>> checking. All functions not included in the Fortran 90 module can
>>> still be invoked from F90 applications, but will fall back to
>>> Fortran-77 style checking (i.e., little/none).
>>> - trivial: Only includes F90-specific functions from MPI-2. This
>>> means overloaded versions of MPI_SIZEOF for all the MPI-supported
>>> F90 intrinsic types.
>>> - small (default): All the functions in "trivial" plus all MPI
>>> functions that take no choice buffers (meaning buffers that are
>>> specified by the user and are of type (void*) in the C bindings --
>>> generally buffers specified for message passing). Hence,
>>> functions like MPI_COMM_RANK are included, but functions like
>>> MPI_SEND are not.
>>> - medium: All the functions in "small" plus all MPI functions that
>>> take one choice buffer (e.g., MPI_SEND, MPI_RECV, ...). All
>>> one-choice-buffer functions have overloaded variants for each of
>>> the MPI-supported Fortran intrinsic types up to the number of
>>> dimensions specified by --with-f90-max-array-dim (default value is
>>> Increasing the size of the F90 module (in order from trivial, small,
>>> and medium) will generally increase the length of time required to
>>> compile user MPI applications. Specifically, "trivial"- and
>>> "small"-sized F90 modules generally allow user MPI applications to
>>> be compiled fairly quickly but lose type safety for all MPI
>>> functions with choice buffers. "medium"-sized F90 modules generally
>>> take longer to compile user applications but provide greater type
>>> safety for MPI functions.
>>> Note that MPI functions with two choice buffers (e.g., MPI_GATHER)
>>> are not currently included in Open MPI's F90 interface. Calls to
>>> these functions will automatically fall through to Open MPI's F77
>>> interface. A "large" size that includes the two choice buffer MPI
>>> functions is possible in future versions of Open MPI.
>>> FWIW, we're arguing^H^H^H^H^H^H^Hdiscussing new Fortran 2003 bindings for
>>> MPI in the MPI-3 Forum right now. We have already addressed the problems
>>> discussed above (F03 now has an equivalent of (void*)), and hope to do a
>>> more minor things as well. There's also discussion of the possibility of
>>> Boost.MPI-like Fortran 2003 MPI library that would take advantage of many
>>> the features of the language, but be a little farther away from the
>>> MPI bindings (see www.boost-org for details about how their nifty C++
>>> library works on top of MPI).
>>> Jeff Squyres
>>> Cisco Systems
>>> users mailing list
>> "In science, there is only physics; all the rest is stamp-collecting."
>> -Ernest Rutherford
>> users mailing list
> Jeff Squyres
> Cisco Systems
> users mailing list
"In science, there is only physics; all the rest is stamp-collecting."